ANALYSIS ON NUCLEAR WEAPONS TESTS FROM 1945 TO 2019¶

INTRODUCTION

According to wiikipedia Nuclear weapons tests are experiments carried out to determine nuclear weapons' effectiveness, yield, and explosive capability. Testing nuclear weapons offers practical information about how the weapons function, how detonations are affected by different conditions, and how personnel, structures, and equipment are affected when subjected to nuclear explosions. However, nuclear testing has often been used as an indicator of scientific and military strength. Many tests have been overtly political in their intention; most nuclear weapons states publicly declared their nuclear status through a nuclear test.

ABOUT THE DATA

This data was collected from https://www.kaggle.com, The data shows the year of each nuclear test by their country. The dataset is reliable, original and comprehensive. The source has their own licence over the dataset. Besides that, the dataset doesn't have any personal information. All the files have consistent columns and each column has the correct type of data. Finally, It would be good to have some updated information about the nuclear weapons tests.

In [1]:
# Nuclear Weapons Tests Analysis with Pandas & Plotly
# We will be using the following Python Libraries: 
# • Pandas
# • Matplotlib
# • Pandas Profiling Report
# • AutoViz
# • Plotly
In [2]:
# We will cover the following chart types:
# • Histogram
# • Area Chart
# • Pie Chart
# • Bar Charts

CONTENTS¶

  • [x] Libraries imports
  • [x] Loading of dataset
  • [x] Dataset exploration
  • [x] Profiling report
  • [x] AutoViz report
  • [x] Data manipulation
  • [x] Data visualization
  • [x] Conclusion

IMPORT OUR LIBRARIES¶

In [3]:
# Imports:
import pandas as pd
import matplotlib.pyplot as plt
import plotly
import plotly.express as px
from pandas_profiling import ProfileReport
from autoviz.AutoViz_Class import AutoViz_Class

Imported v0.1.58. After importing, execute '%matplotlib inline' to display charts in Jupyter.
    AV = AutoViz_Class()
    dfte = AV.AutoViz(filename, sep=',', depVar='', dfte=None, header=0, verbose=1, lowess=False,
               chart_format='svg',max_rows_analyzed=150000,max_cols_analyzed=30, save_plot_dir=None)
Update: verbose=0 displays charts in your local Jupyter notebook.
        verbose=1 additionally provides EDA data cleaning suggestions. It also displays charts.
        verbose=2 does not display charts but saves them in AutoViz_Plots folder in local machine.
        chart_format='bokeh' displays charts in your local Jupyter notebook.
        chart_format='server' displays charts in your browser: one tab for each chart type
        chart_format='html' silently saves interactive HTML files in your local machine

LOADING DATASET

In [4]:
# Now we are going to load our dataset
df = pd.read_csv("nuclear.csv")
In [5]:
df.head()
Out[5]:
country_name year nuclear_weapons_tests
0 China 1945 0
1 China 1946 0
2 China 1947 0
3 China 1948 0
4 China 1949 0

DATASET EXPLORATION¶

In this part I will be exploring the dataset

In [6]:
# Basic info about the DataFrame
df.info()

<class 'pandas.core.frame.DataFrame'>
RangeIndex: 600 entries, 0 to 599
Data columns (total 3 columns):
 #   Column                 Non-Null Count  Dtype 
---  ------                 --------------  ----- 
 0   country_name           600 non-null    object
 1   year                   600 non-null    int64 
 2   nuclear_weapons_tests  600 non-null    int64 
dtypes: int64(2), object(1)
memory usage: 14.2+ KB
In [7]:
# Describe methed
df.describe()
Out[7]:
year nuclear_weapons_tests
count 600.000000 600.000000
mean 1982.000000 3.431667
std 21.666774 9.808789
min 1945.000000 0.000000
25% 1963.000000 0.000000
50% 1982.000000 0.000000
75% 2001.000000 1.000000
max 2019.000000 96.000000
In [8]:
# Get a view of a unique value in column e.g. 'country_name'
df['country_name'].unique()
Out[8]:
array(['China', 'France', 'India', 'North Korea', 'Pakistan', 'Russia',
       'United Kingdom', 'United States'], dtype=object)
In [9]:
# Check for null values
df.isnull()
Out[9]:
country_name year nuclear_weapons_tests
0 False False False
1 False False False
2 False False False
3 False False False
4 False False False
... ... ... ...
595 False False False
596 False False False
597 False False False
598 False False False
599 False False False

600 rows × 3 columns

In [10]:
# NoN count for each column
df.isnull().sum()
Out[10]:
country_name             0
year                     0
nuclear_weapons_tests    0
dtype: int64

PANDAS PROFILING REPORT¶

I will be using pandas-profiling to generate profile reports from my data [nuclear.csv]

In [11]:
# AUTOMATED REPORTS
# Generate pandas profiling report
profile = ProfileReport(df, title = "nuclear.csv Profiling Report")

# To view in Notebook
profile.to_notebook_iframe()

Summarize dataset:   0%|          | 0/5 [00:00<?, ?it/s]
Generate report structure:   0%|          | 0/1 [00:00<?, ?it/s]
Render HTML:   0%|          | 0/1 [00:00<?, ?it/s]

AutoViz REPORT¶

Using AutoViz to perform automatic visualization report on my dataset

In [12]:
# AutoViz report
# Here we would show the AutoViz report 
%matplotlib inline
plt.style.use('classic')
AV = AutoViz_Class()
df_autoviz = AV.AutoViz('nuclear.csv')

Shape of your Data Set loaded: (600, 3)
#######################################################################################
######################## C L A S S I F Y I N G  V A R I A B L E S  ####################
#######################################################################################
Classifying variables in data set...
Data cleaning improvement suggestions. Complete them before proceeding to ML modeling.
  Nuniques dtype Nulls Nullpercent NuniquePercent Value counts Min Data cleaning improvement suggestions
year 75 int64 0 0.000000 12.500000 0
nuclear_weapons_tests 41 int64 0 0.000000 6.833333 0
country_name 8 object 0 0.000000 1.333333 75 
    3 Predictors classified...
        No variables removed since no ID or low-information variables found in data set

<Figure size 1200x0 with 0 Axes>
All Plots done
Time to run AutoViz = 4 seconds 

 ###################### AUTO VISUALIZATION Completed ########################

DATA MANIPULATION¶

Carrying out data manipulation on my dataset in order to gain more insights from my dataset.

First, we find country with the highest and lowest nuclear weapons tests in a year

In [13]:
# Country with the highest Nuclear weapons tests in a year
df.nlargest(1,'nuclear_weapons_tests')
Out[13]:
country_name year nuclear_weapons_tests
542 United States 1962 96
In [14]:
# Country with the smallest Nuclear testsin a year
df.nsmallest(1, 'nuclear_weapons_tests')
Out[14]:
country_name year nuclear_weapons_tests
0 China 1945 0

Finding total sum of nuclear weapons tests per country and grouping them by country names

In [15]:
# Find the total Nuclear weapon tests per country since 1945 to 2019
total =  df.groupby('country_name', as_index=False)['nuclear_weapons_tests'].sum()
total.head()
Out[15]:
country_name nuclear_weapons_tests
0 China 45
1 France 210
2 India 3
3 North Korea 9
4 Pakistan 2
In [16]:
total.tail()
Out[16]:
country_name nuclear_weapons_tests
3 North Korea 9
4 Pakistan 2
5 Russia 715
6 United Kingdom 45
7 United States 1030

DATA VISUALIZATION¶

Visualizing my data

CHART 1: HISTOGRAM SHOWING TOTAL NUCLEAR WEAPONS TEST PER COUNTRY FROM 1945-2019¶

Creating a histogram chart to show the total nuclear weapons test per country from 1945 to 2019

In [17]:
# Create Chart
template_style = 'plotly_white'
fig = px.histogram(total,
                   x ='country_name',
                   y = 'nuclear_weapons_tests',
                   title = '<b>Total Nuclear Weapon Test Per Country 1945-2019</b>',
                   template = template_style,
                   width=800, height=400)
    
# Plot chart
fig.show()

CHART 2: USING AREA CHART TO CHECK FOR TRENDS¶

Checking for trends in nuclear weapons tests from 1945-2019 using Area Chart

In [18]:
# Create Chart
fig = px.area(df,
             x = 'year',
             y = 'nuclear_weapons_tests',
             color = 'country_name',
             template = template_style,
              title = '<b>Area Chart Showing Trends Nuclear Tests</b>',
              width=800, height=400)


# Display Plot
fig.show()

CHART 3: PIE CHART SHOWING PERCENTAGE OF WHOLE¶

I would create a pie chart showing the percentage each country has used in wuclear Weapons testing from 1945-2019

In [19]:
# Create Chart
fig = px.pie(total, 'country_name',
             'nuclear_weapons_tests',
             color = 'nuclear_weapons_tests',
              title = '<b>Pie Chart Nuclear Tests</b>',
             width=800, height=400)


# Display Plot
fig.show()

CHART 4: BAR CHART SHOWING AMOUNT OF YEARLY NUCLEAR WEAPONS TESTS PER EACH COUNTRY¶

In [20]:
# Create a chart
template_style = 'plotly_white'
fig = px.bar(df,
             x = 'year',
             y = 'nuclear_weapons_tests',
             color = 'country_name',
             color_continuous_scale = ['green', 'yellow', 'red'],
             title = '<b>Yearly Nuclear Weapons Tests Per Country 1945-2019</b>',
             template = template_style,)
# Display plot
fig.show()

CHART 5: SCATTER PLOT SHOWING LINEAR CORRELATION BETWEEN YEAR/NUCLEAR WEAPONS TESTS¶

Is there any correlation between Year/Nuclear Weapons Tests & country [Scatter Plot]

In [21]:
fig = px.scatter(df,
             x = 'year',
             y = 'nuclear_weapons_tests',
             color = 'country_name',
             width=800, height=400,
                  template = template_style,
            title = '<b>Scatterplot Year/Nuclear Weapons Test</b>')

fig.show()

CONCLUSION¶

In this conclusion I would talk about the noticed trends in my insights¶

Since the end of world war 2, various nuclear weapons tests has been carried out by some countries, which is the basis of my analysis and insights. This analysis gave me some important insights on how countries has been testing nuclear weapons for years. What I found out was that;

  • The United States carried out the highest nuclear weapons test from 1945 to 2019 with a total of 1030 tests.
  • Russia had the second highest nuclear weapons tests from 1945 to 2019 with a total of 715 tests.
  • Pakistan had the lowest nuclear weapons test with a total of 2 tests.
  • The United States had the highest nuclear weapons tests in a calendar year with a total of 96 tests in 1962.
  • China had no nuclear weapons tests in 1945 but had a total of 45 nuclear weapons tests since 1945 to 2019.
  • The United States has tested about 50% of the world's nuclear weapons test from 1945 to 2019.
  • Within 1955 to 1956 the United States has carried out about 36 nuclear weapons tests with each of the year having 18 tests each.
  • The United States had a consistent nuclear weapons tests from 1971 to 1992, making it 21 years of consistent nuclear weapons testing.
  • France also had a consistent nuclear weapons testing from 1970 to 1991 making it 21 years of consistent nuclear weapons testing.
  • Russia's consistent nuclear weapons testing occurred between 1964 to 1985 making it also 21 years of consistent nuclear weapons testing.
  • The U.S, Russia and France had 21 years each of consistent nuclear weapons testing.
  • Russia has tested 34.7% of the world's nuclear weapons tests, while France has tested 10.2% of the world's nuclear weapons tests, making them one of the top 3 nuclear weapons testers.
  • This report shows that the United states, Russia and France has carried out 94.9% of the world's nuclear weapons tests from 1945 to 2019.
In [ ]: